Publishing & AI Agents
LivePropertyNews.io - Autonomous AI Publishing Platform
Built a fully autonomous property news platform that sources, rewrites, and publishes 16+ articles daily across 8 categories. The platform uses an AI-powered content pipeline - RSS feeds are ingested, processed through Claude, and published to a custom Next.js frontend with zero human intervention. Includes city-level landing pages with live market data, SEO infrastructure, and an admin dashboard for monitoring pipeline health.
Visit live siteThe challenge
What we were solving
Property professionals and investors need to stay on top of market news across multiple categories and regions, but no single source covers the full picture. Existing property news sites rely on manual editorial teams, making them expensive to run and slow to publish. The client wanted a platform that could deliver comprehensive, timely UK property news coverage without the overhead of a traditional newsroom.
Our approach
How we built it
We designed a fully autonomous content pipeline that monitors 17 RSS feeds from leading property news sources. Each run, the system identifies the most relevant stories, processes them through Claude to produce original 800-1,200 word articles, applies category and city tagging, generates SEO metadata, and publishes directly to Supabase. The Next.js frontend pulls content in real-time, with city landing pages enriched by live market statistics. A GitHub Actions workflow triggers the pipeline twice daily, and an admin dashboard gives complete visibility into system health and subscriber activity.
Results
What we delivered
16+ articles published daily, fully autonomous
8 content categories covering the UK property market
7 city landing pages with live market data and statistics
Automated twice-daily publishing via GitHub Actions
Full SEO infrastructure with dynamic OG images and JSON-LD
Admin dashboard for pipeline monitoring and subscriber management
Tech stack
Built with
Deep dive
The full picture
PropertyNews.io represents our most ambitious autonomous publishing build to date. The platform monitors 17 RSS feeds from major UK property news sources, identifies the most relevant stories, and processes them through a Claude-powered pipeline that rewrites each article into a unique 800-1,200 word piece with proper attribution, SEO metadata, and category tagging.
The frontend is a fully custom Next.js application with 7 city landing pages - each pulling live market data including average property prices, growth rates, and regional statistics. The site includes 8 content categories covering housing, commercial, rental, investment, development, mortgages, policy, and strategy - giving readers comprehensive coverage of the entire UK property market.
The admin dashboard provides full visibility into pipeline health, subscriber management, and content performance. A twice-daily GitHub Actions workflow triggers the pipeline at 7am and 1pm UTC, ensuring fresh content throughout the day without any manual intervention. The entire system - from RSS ingestion to published article - runs without human involvement.
Got a similar challenge?
We build AI agents, automation workflows, and custom tools that solve real business problems. If this case study resonated, let's talk about what we can build for you.
Start a conversation